A data collection and analysis engine built into an application installed on a client device may be employed to collect, analyze, and report application data desired from the client device to a service provider associated with the application. However, in conventional systems, data collection performed by the data collection and analysis engine may be constrained by one or more rules defined by a developer to collect an event in a static manner. Often, out of concern for data volume the event may generate, as well as device resources that the data collection and analysis engine may consume, the developer may design the rules to fit a lowest-end device resource profile in order to avoid performance issues on low-end devices.
As such, the developer may have to design many different rules for collecting a same event in order to fit resource profiles for a multitude of device types, and with the sheer number of devices employed throughout the world, the time and cost to design data collection rules for various events for each device type would be excessive. Accordingly, methods for data collection could use improvements such that implementation of dynamically variable data collection patterns may be enabled, where the data collections patterns correspond to resources for a particular, target device.